Skip to content

feat: Unify keyword and semantic search into 'Search My Library' #37

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
amanvatss wants to merge 1 commit into
base: main
Choose a base branch
from
Open

feat: Unify keyword and semantic search into 'Search My Library' #37

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
285 changes: 144 additions & 141 deletions streamlit_app.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -548,13 +548,13 @@ def to_date(iso_str: str) -> Optional[datetime.date]:
with st.sidebar:
st.markdown('<div class="sidebar-title">Navigation</div>', unsafe_allow_html=True)

# === MODIFIED: Updated the sidebar options ===
option = st.radio(
label="Choose a section",
options=[
"Search arXiv Papers",
"Query Stored Summaries",
"Upload & Summarize PDF",
"Semantic Search (FAISS)",
"Search My Library", # <-- This is the new unified page
"History"
],
key="main_menu",
Expand Down Expand Up @@ -663,7 +663,7 @@ def to_date(iso_str: str) -> Optional[datetime.date]:
else:
st.error("Failed to fetch paper from URL. Please check the URL and try again")
st.info("Supported formats: arXiv URLs (abs or pdf) and direct PDF links")

else:
with st.spinner("Fetching and summarizing..."):
data = fetch_and_summarize(keyword_input)
Expand Down Expand Up @@ -718,83 +718,16 @@ def to_date(iso_str: str) -> Optional[datetime.date]:
)
else:
st.warning("No results found")

except Exception as e:
st.error(f"Error processing: {str(e)}")
st.info("Supported formats: arXiv URLs (abs or pdf) and direct PDF links")

st.markdown('</div>', unsafe_allow_html=True)

# ---------------------------------------------------------------------
# Section: Query Stored Summaries
# === MODIFIED: "Query Stored Summaries" block DELETED ===
# ---------------------------------------------------------------------
elif option == "Query Stored Summaries":
st.markdown('<div class="section-wrapper">', unsafe_allow_html=True)
st.markdown('<div class="section-title">Query Stored Summaries</div>', unsafe_allow_html=True)

query_input = st.text_input(
"Search stored summaries",
placeholder="e.g., contrastive learning"
)

if st.button("Run Keyword Search"):
if not query_input or query_input.strip() == "":
st.warning("Please enter a search query")
else:
with st.spinner("Searching database..."):
try:
results = query_db(query_input)

if not results:
st.info("No matching summaries found")
else:
st.success(f"Found {len(results)} results")
for idx, item in enumerate(results):
with st.expander(f"{item.get('title', 'Untitled')}", expanded=(idx == 0)):
st.markdown("**Summary:**")
st.write(item.get('summary', 'No summary available'))

# Download buttons
download_col1, download_col2, download_col3 = st.columns([1, 1, 1])
summary_text = item.get('summary', '')
md_bytes = f"# {item.get('title', '')}\n\n{summary_text}".encode("utf-8")
txt_bytes = summary_text.encode("utf-8")
pdf_buf = generate_pdf_from_text(
title=item.get('title', 'Summary'),
metadata={"id": item.get('id', '')},
body_text=summary_text,
annotations=item.get('annotations', '')
)

with download_col1:
st.download_button(
label="TXT",
data=txt_bytes,
file_name=f"{safe_filename(item.get('title', 'summary'))}.txt",
mime="text/plain",
key=f"query_txt_{idx}"
)
with download_col2:
st.download_button(
label="MD",
data=md_bytes,
file_name=f"{safe_filename(item.get('title', 'summary'))}.md",
mime="text/markdown",
key=f"query_md_{idx}"
)
with download_col3:
st.download_button(
label="PDF",
data=pdf_buf,
file_name=f"{safe_filename(item.get('title', 'summary'))}.pdf",
mime="application/pdf",
key=f"query_pdf_{idx}"
)

except Exception as e:
st.error(f"Search failed: {str(e)}")

st.markdown('</div>', unsafe_allow_html=True)

# ---------------------------------------------------------------------
# Section: Upload & Summarize PDF
Expand Down Expand Up @@ -903,86 +836,156 @@ def to_date(iso_str: str) -> Optional[datetime.date]:
st.markdown('</div>', unsafe_allow_html=True)

# ---------------------------------------------------------------------
# Section: Semantic Search (FAISS)
# === MODIFIED: "Semantic Search (FAISS)" block DELETED ===
# ---------------------------------------------------------------------

# ---------------------------------------------------------------------
# === MODIFIED: New "Search My Library" Page ADDED ===
# ---------------------------------------------------------------------
elif option == "Semantic Search (FAISS)":
elif option == "Search My Library":
st.markdown('<div class="section-wrapper">', unsafe_allow_html=True)
st.markdown('<div class="section-title">Semantic Search (FAISS)</div>', unsafe_allow_html=True)

if st.button("Rebuild Index"):
if DEMO_MODE:
st.info("Index building is simulated in Demo Mode. Load the demo instead")
else:
with st.spinner("Rebuilding semantic vector index..."):
try:
build_index()
st.success("Index rebuilt successfully")
except Exception as e:
st.error(f"Failed to rebuild index: {e}")
st.markdown('<div class="section-title">Search My Library</div>', unsafe_allow_html=True)

# --- Logic moved from old "Semantic Search" page ---
with st.expander("🗂️ Index Management"):
if st.button("Rebuild Index"):
if DEMO_MODE:
st.info("Index building is simulated in Demo Mode. Load the demo instead")
else:
with st.spinner("Rebuilding semantic vector index..."):
try:
build_index()
st.success("Index rebuilt successfully")
except Exception as e:
st.error(f"Failed to rebuild index: {e}")

semantic_query = st.text_input(
"Enter a semantic query",
placeholder="e.g., visual prompt tuning in robotics"
# --- New Unified Search UI ---
search_mode = st.radio(
"Select search mode:",
("Keyword", "Semantic"),
horizontal=True
)

query_input = st.text_input(
"Enter your search query",
placeholder="e.g., 'contrastive learning' (Keyword) or 'robotics visual tuning' (Semantic)"
)

if st.button("Search with FAISS"):
if not semantic_query or semantic_query.strip() == "":
st.warning("Please enter a semantic query")
if st.button("🔍 Search Library"):
if not query_input or query_input.strip() == "":
st.warning("Please enter a search query")
else:
with st.spinner("Running semantic search..."):
try:
results = search_similar(semantic_query)

if not results:
st.info("No similar results found")
else:
st.success(f"Found {len(results)} similar items")
for idx, item in enumerate(results):
with st.expander(f"{item.get('title', 'Untitled')}", expanded=(idx == 0)):
st.markdown("**Summary:**")
st.write(item.get('summary', 'No summary available'))
if search_mode == "Keyword":
# --- This is the logic from the old "Query Stored Summaries" page ---
with st.spinner("Searching database (Keyword)..."):
try:
results = query_db(query_input)

if not results:
st.info("No matching summaries found")
else:
st.success(f"Found {len(results)} results")
for idx, item in enumerate(results):
with st.expander(f"{item.get('title', 'Untitled')}", expanded=(idx == 0)):
st.markdown("**Summary:**")
st.write(item.get('summary', 'No summary available'))

download_col1, download_col2, download_col3 = st.columns([1, 1, 1])
summary_text = item.get('summary', '')
md_bytes = f"# {item.get('title', '')}\n\n{summary_text}".encode("utf-8")
txt_bytes = summary_text.encode("utf-8")
pdf_buf = generate_pdf_from_text(
title=item.get('title', 'Summary'),
metadata={"id": item.get('id', '')},
body_text=summary_text,
annotations=item.get('annotations', '')
)

with download_col1:
st.download_button(
label="TXT",
data=txt_bytes,
file_name=f"{safe_filename(item.get('title', 'summary'))}.txt",
mime="text/plain",
key=f"faiss_txt_{idx}"
)
with download_col2:
st.download_button(
label="MD",
data=md_bytes,
file_name=f"{safe_filename(item.get('title', 'summary'))}.md",
mime="text/markdown",
key=f"faiss_md_{idx}"
# --- Download button logic from old page ---
download_col1, download_col2, download_col3 = st.columns([1, 1, 1])
summary_text = item.get('summary', '')
md_bytes = f"# {item.get('title', '')}\n\n{summary_text}".encode("utf-8")
txt_bytes = summary_text.encode("utf-8")
pdf_buf = generate_pdf_from_text(
title=item.get('title', 'Summary'),
metadata={"id": item.get('id', '')},
body_text=summary_text,
annotations=item.get('annotations', '')
)
with download_col3:
st.download_button(
label="PDF",
data=pdf_buf,
file_name=f"{safe_filename(item.get('title', 'summary'))}.pdf",
mime="application/pdf",
key=f"faiss_pdf_{idx}"

with download_col1:
st.download_button(
label="TXT",
data=txt_bytes,
file_name=f"{safe_filename(item.get('title', 'summary'))}.txt",
mime="text/plain",
key=f"lib_query_txt_{idx}"
)
with download_col2:
st.download_button(
label="MD",
data=md_bytes,
file_name=f"{safe_filename(item.get('title', 'summary'))}.md",
mime="text/markdown",
key=f"lib_query_md_{idx}"
)
with download_col3:
st.download_button(
label="PDF",
data=pdf_buf,
file_name=f"{safe_filename(item.get('title', 'summary'))}.pdf",
mime="application/pdf",
key=f"lib_query_pdf_{idx}"
)
except Exception as e:
st.error(f"Search failed: {str(e)}")

elif search_mode == "Semantic":
# --- This is the logic from the old "Semantic Search (FAISS)" page ---
with st.spinner("Running semantic search..."):
try:
results = search_similar(query_input)
Copy link

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Inconsistent indentation detected. This line appears to have extra indentation compared to the surrounding code block. The indentation should align with other statements at the same nesting level (line 882 uses results = query_db(query_input) with consistent indentation).

Suggested change
results = search_similar(query_input)
results = search_similar(query_input)

Copilot uses AI. Check for mistakes.

if not results:
st.info("No similar results found")
else:
st.success(f"Found {len(results)} similar items")
for idx, item in enumerate(results):
with st.expander(f"{item.get('title', 'Untitled')}", expanded=(idx == 0)):
st.markdown("**Summary:**")
st.write(item.get('summary', 'No summary available'))

# --- Download button logic from old page ---
download_col1, download_col2, download_col3 = st.columns([1, 1, 1])
summary_text = item.get('summary', '')
md_bytes = f"# {item.get('title', '')}\n\n{summary_text}".encode("utf-8")
txt_bytes = summary_text.encode("utf-8")
pdf_buf = generate_pdf_from_text(
title=item.get('title', 'Summary'),
metadata={"id": item.get('id', '')},
body_text=summary_text,
annotations=item.get('annotations', '')
)

with download_col1:
st.download_button(
label="TXT",
data=txt_bytes,
file_name=f"{safe_filename(item.get('title', 'summary'))}.txt",
mime="text/plain",
key=f"lib_faiss_txt_{idx}"
)
with download_col2:
st.download_button(
label="MD",
data=md_bytes,
file_name=f"{safe_filename(item.get('title', 'summary'))}.md",
mime="text/markdown",
key=f"lib_faiss_md_{idx}"
)
with download_col3:
st.download_button(
label="PDF",
data=pdf_buf,
file_name=f"{safe_filename(item.get('title', 'summary'))}.pdf",
mime="application/pdf",
key=f"lib_faiss_pdf_{idx}"
)
except Exception as e:
st.error(f"Semantic search failed: {e}")

except Exception as e:
st.error(f"Semantic search failed: {e}")

st.markdown('</div>', unsafe_allow_html=True)


# ---------------------------------------------------------------------
# Section: History
# ---------------------------------------------------------------------
Expand Down Expand Up @@ -1153,4 +1156,4 @@ def to_date(iso_str: str) -> Optional[datetime.date]:
key=f"history_pdf_{idx}"
)

st.markdown('</div>', unsafe_allow_html=True)
st.markdown('</div>', unsafe_allow_html=True)
Loading